-
Notifications
You must be signed in to change notification settings - Fork 2
Expand file tree
/
Copy pathstreamio.py
More file actions
1244 lines (1073 loc) · 40.7 KB
/
streamio.py
File metadata and controls
1244 lines (1073 loc) · 40.7 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
"""New standard I/O library.
Based on sio.py from Guido van Rossum.
- This module contains various stream classes which provide a subset of the
classic Python I/O API: read(n), write(s), tell(), seek(offset, whence=0),
readall(), readline(), truncate(size), flush(), close(), peek(),
flushable(), try_to_find_file_descriptor().
- This is not for general usage:
* read(n) may return less than n bytes, just like os.read().
* some other methods also have no default parameters.
* close() should be called exactly once and no further operations performed;
there is no __del__() closing the stream for you.
* some methods may raise MyNotImplementedError.
* peek() returns some (or no) characters that have already been read ahead.
* flushable() returns True/False if flushing that stream is useful/pointless.
- A 'basis stream' provides I/O using a low-level API, like the os, mmap or
socket modules.
- A 'filtering stream' builds on top of another stream. There are filtering
streams for universal newline translation, for unicode translation, and
for buffering.
You typically take a basis stream, place zero or more filtering
streams on top of it, and then top it off with an input-buffering and/or
an outout-buffering stream.
"""
# File offsets are all 'r_longlong', but a single read or write cannot
# transfer more data that fits in an RPython 'int' (because that would not
# fit in a single string anyway). This module needs to be careful about
# where r_longlong values end up: as argument to seek() and truncate() and
# return value of tell(), but not as argument to read().
import os, sys, errno
from rpython.rlib.objectmodel import specialize, we_are_translated, not_rpython
from rpython.rlib.rarithmetic import r_longlong, intmask
from rpython.rlib import rposix, nonconst, _rsocket_rffi as _c
from rpython.rlib.rstring import StringBuilder
from os import O_RDONLY, O_WRONLY, O_RDWR, O_CREAT, O_TRUNC, O_APPEND
O_BINARY = getattr(os, "O_BINARY", 0)
# (basemode, plus)
OS_MODE = {('r', False): O_RDONLY,
('r', True): O_RDWR,
('w', False): O_WRONLY | O_CREAT | O_TRUNC,
('w', True): O_RDWR | O_CREAT | O_TRUNC,
('a', False): O_WRONLY | O_CREAT | O_APPEND,
('a', True): O_RDWR | O_CREAT | O_APPEND,
}
class MyNotImplementedError(Exception):
"""Catching NotImplementedError is not RPython, so we use this custom class
instead of it
"""
# ____________________________________________________________
def replace_crlf_with_lf(s):
substrings = s.split("\r")
result = [substrings[0]]
for substring in substrings[1:]:
if not substring:
result.append("")
elif substring[0] == "\n":
result.append(substring[1:])
else:
result.append(substring)
return "\n".join(result)
def replace_char_with_str(string, c, s):
return s.join(string.split(c))
@specialize.argtype(0)
def open_file_as_stream(path, mode="r", buffering=-1, signal_checker=None):
os_flags, universal, reading, writing, basemode, binary = decode_mode(mode)
stream = open_path_helper(path, os_flags, basemode == "a", signal_checker)
return construct_stream_tower(stream, buffering, universal, reading,
writing, binary)
def _setfd_binary(fd):
pass
if hasattr(_c, 'fcntl'):
def _check_fd_mode(fd, reading, writing):
flags = intmask(_c.fcntl(fd, _c.F_GETFL, 0))
if flags & _c.O_RDWR:
return
elif flags & _c.O_WRONLY:
if not reading:
return
else: # O_RDONLY
if not writing:
return
raise OSError(22, "Invalid argument")
else:
def _check_fd_mode(fd, reading, writing):
# XXX
pass
def fdopen_as_stream(fd, mode, buffering=-1, signal_checker=None):
os_flags, universal, reading, writing, basemode, binary = decode_mode(mode)
_check_fd_mode(fd, reading, writing)
_setfd_binary(fd)
stream = DiskFile(fd, signal_checker)
return construct_stream_tower(stream, buffering, universal, reading,
writing, binary)
@specialize.argtype(0)
def open_path_helper(path, os_flags, append, signal_checker=None):
# XXX for now always return DiskFile
fd = rposix.open(path, os_flags, 0666)
if append:
try:
os.lseek(fd, 0, 2)
except OSError:
# XXX does this pass make sense?
pass
return DiskFile(fd, signal_checker)
def decode_mode(mode):
if mode[0] == 'U':
mode = 'r' + mode
basemode = mode[0] # 'r', 'w' or 'a'
plus = False
universal = False
binary = False
for c in mode[1:]:
if c == '+':
plus = True
elif c == 'U':
universal = True
elif c == 'b':
binary = True
else:
break
flag = OS_MODE[basemode, plus]
flag |= O_BINARY
reading = basemode == 'r' or plus
writing = basemode != 'r' or plus
return flag, universal, reading, writing, basemode, binary
def construct_stream_tower(stream, buffering, universal, reading, writing,
binary):
if buffering == 0: # no buffering
pass
elif buffering == 1: # line-buffering
if writing:
stream = LineBufferingOutputStream(stream)
if reading:
stream = BufferingInputStream(stream)
else: # default or explicit buffer sizes
if buffering is not None and buffering < 0:
buffering = -1
if writing:
stream = BufferingOutputStream(stream, buffering)
if reading:
stream = BufferingInputStream(stream, buffering)
if universal: # Wants universal newlines
if writing and os.linesep != '\n':
stream = TextOutputFilter(stream)
if reading:
stream = TextInputFilter(stream)
elif not binary and os.linesep == '\r\n':
stream = TextCRLFFilter(stream)
if nonconst.NonConstant(False):
stream.flush_buffers() # annotation workaround for untranslated tests
return stream
class StreamError(Exception):
def __init__(self, message):
self.message = message
StreamErrors = (OSError, StreamError) # errors that can generally be raised
if sys.platform == "win32":
from rpython.rlib.rwin32 import BOOL, HANDLE, get_osfhandle
from rpython.rlib.rwin32 import GetLastError_saved
from rpython.translator.tool.cbuild import ExternalCompilationInfo
from rpython.rtyper.lltypesystem import rffi
_eci = ExternalCompilationInfo()
_setmode = rffi.llexternal('_setmode', [rffi.INT, rffi.INT], rffi.INT,
compilation_info=_eci)
SetEndOfFile = rffi.llexternal('SetEndOfFile', [HANDLE], BOOL,
compilation_info=_eci,
save_err=rffi.RFFI_SAVE_LASTERROR)
def _setfd_binary(fd):
# Allow this to succeed on invalid fd's
if rposix.is_valid_fd(fd):
_setmode(fd, os.O_BINARY)
def ftruncate_win32(fd, size):
curpos = os.lseek(fd, 0, 1)
try:
# move to the position to be truncated
os.lseek(fd, size, 0)
# Truncate. Note that this may grow the file!
handle = get_osfhandle(fd)
if not SetEndOfFile(handle):
raise OSError(GetLastError_saved(),
"Could not truncate file")
finally:
# we restore the file pointer position in any case
os.lseek(fd, curpos, 0)
class Stream(object):
"""Base class for streams. Provides a default implementation of
some methods."""
def read(self, n):
raise MyNotImplementedError
def write(self, data):
raise MyNotImplementedError
def tell(self):
raise MyNotImplementedError
def seek(self, offset, whence):
raise MyNotImplementedError
def readall(self):
bufsize = 8192
result = []
while True:
try:
data = self.read(bufsize)
except OSError:
# like CPython < 3.4, partial results followed by an error
# are returned as data
if not result:
raise
break
if not data:
break
result.append(data)
if bufsize < 4194304: # 4 Megs
bufsize <<= 1
return ''.join(result)
def readline(self):
# very inefficient unless there is a peek()
result = []
while True:
# "peeks" on the underlying stream to see how many characters
# we can safely read without reading past an end-of-line
startindex, peeked = self.peek()
assert 0 <= startindex <= len(peeked)
pn = peeked.find("\n", startindex)
if pn < 0:
pn = len(peeked)
c = self.read(pn - startindex + 1)
if not c:
break
result.append(c)
if c.endswith('\n'):
break
return ''.join(result)
def truncate(self, size):
raise MyNotImplementedError
def flush_buffers(self):
pass
def flush(self):
pass
def flushable(self):
return False
def close(self):
self.close1(True)
def close1(self, closefileno):
pass
def peek(self):
return (0, '')
def count_buffered_bytes(self):
pos, buf = self.peek()
return len(buf) - pos
def try_to_find_file_descriptor(self):
return -1
def getnewlines(self):
return 0
class DiskFile(Stream):
"""Standard I/O basis stream using os.open/close/read/write/lseek"""
def __init__(self, fd, signal_checker=None):
self.fd = fd
self.signal_checker = signal_checker
def seek(self, offset, whence):
os.lseek(self.fd, offset, whence)
def tell(self):
return os.lseek(self.fd, 0, 1)
def read(self, n):
assert isinstance(n, int)
while True:
try:
return os.read(self.fd, n)
except OSError as e:
if e.errno != errno.EINTR:
raise
if self.signal_checker is not None:
self.signal_checker()
# else try again
def readline(self):
# mostly inefficient, but not as laughably bad as with the default
# readline() from Stream
result = StringBuilder()
while True:
try:
c = os.read(self.fd, 1)
except OSError as e:
if e.errno != errno.EINTR:
raise
if self.signal_checker is not None:
self.signal_checker()
continue # try again
if not c:
break
c = c[0]
result.append(c)
if c == '\n':
break
return result.build()
def write(self, data):
while data:
try:
n = os.write(self.fd, data)
except OSError as e:
if e.errno != errno.EINTR:
raise
if self.signal_checker is not None:
self.signal_checker()
else:
data = data[n:]
def close1(self, closefileno):
if closefileno:
os.close(self.fd)
if sys.platform == "win32":
def truncate(self, size):
ftruncate_win32(self.fd, size)
else:
def truncate(self, size):
# Note: for consistency, in translated programs a failing
# os.ftruncate() raises OSError. However, on top of
# CPython, we get an IOError. As it is (as far as I know)
# the only place that have this behavior, we just convert it
# to an OSError instead of adding IOError to StreamErrors.
if we_are_translated():
os.ftruncate(self.fd, size)
else:
try:
os.ftruncate(self.fd, size)
except IOError as e:
raise OSError(*e.args)
def try_to_find_file_descriptor(self):
return self.fd
# next class is not RPython
class MMapFile(Stream):
"""Standard I/O basis stream using mmap."""
@not_rpython
def __init__(self, fd, mmapaccess):
self.fd = fd
self.access = mmapaccess
self.pos = 0
self.remapfile()
def remapfile(self):
import mmap
size = os.fstat(self.fd).st_size
self.mm = mmap.mmap(self.fd, size, access=self.access)
def close1(self, closefileno):
self.mm.close()
if closefileno:
os.close(self.fd)
def tell(self):
return self.pos
def seek(self, offset, whence):
if whence == 0:
self.pos = max(0, offset)
elif whence == 1:
self.pos = max(0, self.pos + offset)
elif whence == 2:
self.pos = max(0, self.mm.size() + offset)
else:
raise StreamError("seek(): whence must be 0, 1 or 2")
def readall(self):
filesize = self.mm.size() # Actual file size, may be more than mapped
n = filesize - self.pos
data = self.mm[self.pos:]
if len(data) < n:
del data
# File grew since opened; remap to get the new data
self.remapfile()
data = self.mm[self.pos:]
self.pos += len(data)
return data
def read(self, n):
assert isinstance(n, int)
end = self.pos + n
data = self.mm[self.pos:end]
if not data:
# is there more data to read?
filesize = self.mm.size() #Actual file size, may be more than mapped
if filesize > self.pos:
# File grew since opened; remap to get the new data
self.remapfile()
data = self.mm[self.pos:end]
self.pos += len(data)
return data
def readline(self):
hit = self.mm.find("\n", self.pos) + 1
if not hit:
# is there more data to read?
filesize = self.mm.size() #Actual file size, may be more than mapped
if filesize > len(self.mm):
# File grew since opened; remap to get the new data
self.remapfile()
hit = self.mm.find("\n", self.pos) + 1
if hit:
# Got a whole line
data = self.mm[self.pos:hit]
self.pos = hit
else:
# Read whatever we've got -- may be empty
data = self.mm[self.pos:]
self.pos += len(data)
return data
def write(self, data):
end = self.pos + len(data)
try:
self.mm[self.pos:end] = data
# This can raise IndexError on Windows, ValueError on Unix
except (IndexError, ValueError):
# XXX On Unix, this resize() call doesn't work
self.mm.resize(end)
self.mm[self.pos:end] = data
self.pos = end
def flush(self):
self.mm.flush()
def flushable(self):
import mmap
return self.access == mmap.ACCESS_WRITE
def try_to_find_file_descriptor(self):
return self.fd
# ____________________________________________________________
STREAM_METHODS = dict([
("read", [int]),
("write", [str]),
("tell", []),
("seek", [r_longlong, int]),
("readall", []),
("readline", []),
("truncate", [r_longlong]),
("flush", []),
("flushable", []),
("close1", [int]),
("peek", []),
("try_to_find_file_descriptor", []),
("getnewlines", []),
])
def PassThrough(meth_name, flush_buffers):
if meth_name in STREAM_METHODS:
signature = STREAM_METHODS[meth_name]
args = ", ".join(["v%s" % (i, ) for i in range(len(signature))])
else:
assert 0, "not a good idea"
args = "*args"
if flush_buffers:
code = """def %s(self, %s):
self.flush_buffers()
return self.base.%s(%s)
"""
else:
code = """def %s(self, %s):
return self.base.%s(%s)
"""
d = {}
exec code % (meth_name, args, meth_name, args) in d
return d[meth_name]
def offset2int(offset):
intoffset = intmask(offset)
if intoffset != offset:
raise StreamError("seek() from a non-seekable source:"
" this would read and discard more"
" than sys.maxint bytes")
return intoffset
class BufferingInputStream(Stream):
"""Standard buffering input stream.
This, and BufferingOutputStream if needed, are typically at the top of
the stack of streams.
"""
bigsize = 2**19 # Half a Meg
bufsize = 2**13 # 8 K
def __init__(self, base, bufsize=-1):
self.base = base
self.do_read = base.read # function to fill buffer some more
self.do_tell = base.tell # return a byte offset
self.do_seek = base.seek # seek to a byte offset
if bufsize == -1: # Get default from the class
bufsize = self.bufsize
self.bufsize = bufsize # buffer size (hint only)
self.buf = "" # raw data
self.pos = 0
def flush_buffers(self):
if self.buf:
try:
self.do_seek(self.pos - len(self.buf), 1)
except (MyNotImplementedError, OSError):
pass
else:
self.buf = ""
self.pos = 0
def tell(self):
tellpos = self.do_tell() # This may fail
# Best-effort: to avoid extra system calls to tell() all the
# time, and a more complicated logic in this class, we can
# only assume that nobody changed the underlying file
# descriptor position while we have buffered data. If they
# do, we might get bogus results here (and the following
# read() will still return the data cached at the old
# position). Just make sure that we don't fail an assert.
offset = len(self.buf) - self.pos
if tellpos < offset:
# bug! someone changed the fd position under our feet,
# and moved it at or very close to the beginning of the
# file, so that we have more buffered data than the
# current offset.
self.buf = ""
self.pos = 0
offset = 0
return tellpos - offset
def seek(self, offset, whence):
# This may fail on the do_seek() or on the tell() call.
# But it won't depend on either on a relative forward seek.
# Nor on a seek to the very end.
if whence == 0 or whence == 1:
if whence == 0:
difpos = offset - self.tell() # may clean up self.buf/self.pos
else:
difpos = offset
currentsize = len(self.buf) - self.pos
if -self.pos <= difpos <= currentsize:
self.pos += intmask(difpos)
return
if whence == 1:
offset -= currentsize
try:
self.do_seek(offset, whence)
except MyNotImplementedError:
self.buf = ""
self.pos = 0
if difpos < 0:
raise
if whence == 0:
offset = difpos - currentsize
intoffset = offset2int(offset)
self.read(intoffset)
else:
self.buf = ""
self.pos = 0
return
if whence == 2:
try:
self.do_seek(offset, 2)
except MyNotImplementedError:
pass
else:
self.pos = 0
self.buf = ""
return
# Skip relative to EOF by reading and saving only just as
# much as needed
intoffset = offset2int(offset)
pos = self.pos
assert pos >= 0
buffers = [self.buf[pos:]]
total = len(buffers[0])
self.buf = ""
self.pos = 0
while 1:
data = self.do_read(self.bufsize)
if not data:
break
buffers.append(data)
total += len(data)
while buffers and total >= len(buffers[0]) - intoffset:
total -= len(buffers[0])
del buffers[0]
cutoff = total + intoffset
if cutoff < 0:
raise StreamError("cannot seek back")
if buffers:
assert cutoff >= 0
buffers[0] = buffers[0][cutoff:]
self.buf = "".join(buffers)
return
raise StreamError("whence should be 0, 1 or 2")
def readall(self):
pos = self.pos
assert pos >= 0
if self.buf:
chunks = [self.buf[pos:]]
else:
chunks = []
self.buf = ""
self.pos = 0
bufsize = self.bufsize
while 1:
try:
data = self.do_read(bufsize)
except OSError as o:
# like CPython < 3.4, partial results followed by an error
# are returned as data
if not chunks:
raise
break
if not data:
break
chunks.append(data)
bufsize = min(bufsize*2, self.bigsize)
return "".join(chunks)
def read(self, n=-1):
assert isinstance(n, int)
if n < 0:
return self.readall()
currentsize = len(self.buf) - self.pos
start = self.pos
assert start >= 0
if n <= currentsize:
stop = start + n
assert stop >= 0
result = self.buf[start:stop]
self.pos += n
return result
else:
chunks = [self.buf[start:]]
while 1:
self.buf = self.do_read(self.bufsize)
if not self.buf:
self.pos = 0
break
currentsize += len(self.buf)
if currentsize >= n:
self.pos = len(self.buf) - (currentsize - n)
stop = self.pos
assert stop >= 0
chunks.append(self.buf[:stop])
break
buf = self.buf
assert buf is not None
chunks.append(buf)
return ''.join(chunks)
def readline(self):
pos = self.pos
assert pos >= 0
i = self.buf.find("\n", pos)
start = self.pos
assert start >= 0
if i >= 0: # new line found
i += 1
result = self.buf[start:i]
self.pos = i
return result
temp = self.buf[start:]
# read one buffer and most of the time a new line will be found
self.buf = self.do_read(self.bufsize)
i = self.buf.find("\n")
if i >= 0: # new line found
i += 1
result = temp + self.buf[:i]
self.pos = i
return result
if not self.buf:
self.pos = 0
return temp
# need to keep getting data until we find a new line
chunks = [temp, self.buf]
while 1:
self.buf = self.do_read(self.bufsize)
if not self.buf:
self.pos = 0
break
i = self.buf.find("\n")
if i >= 0:
i += 1
chunks.append(self.buf[:i])
self.pos = i
break
chunks.append(self.buf)
return "".join(chunks)
def peek(self):
return (self.pos, self.buf)
write = PassThrough("write", flush_buffers=True)
truncate = PassThrough("truncate", flush_buffers=True)
flush = PassThrough("flush", flush_buffers=True)
flushable = PassThrough("flushable", flush_buffers=False)
close1 = PassThrough("close1", flush_buffers=False)
try_to_find_file_descriptor = PassThrough("try_to_find_file_descriptor",
flush_buffers=False)
class BufferingOutputStream(Stream):
"""Standard buffering output stream.
This, and BufferingInputStream if needed, are typically at the top of
the stack of streams.
"""
bigsize = 2**19 # Half a Meg
bufsize = 2**13 # 8 K
def __init__(self, base, bufsize=-1):
self.base = base
self.do_tell = base.tell # return a byte offset
if bufsize == -1: # Get default from the class
bufsize = self.bufsize
self.bufsize = bufsize # buffer size (hint only)
self.buf = []
self.buflen = 0
self.error = False
def do_write(self, data):
try:
self.base.write(data)
except:
self.error = True
raise
def flush_buffers(self):
if self.buf and not self.error:
self.do_write(''.join(self.buf))
self.buf = []
self.buflen = 0
def tell(self):
return self.do_tell() + self.buflen
def write(self, data):
self.error = False
buflen = self.buflen
datalen = len(data)
if datalen + buflen < self.bufsize:
self.buf.append(data)
self.buflen += datalen
elif buflen:
self.buf.append(data)
self.do_write(''.join(self.buf))
self.buf = []
self.buflen = 0
else:
self.do_write(data)
read = PassThrough("read", flush_buffers=True)
readall = PassThrough("readall", flush_buffers=True)
readline = PassThrough("readline", flush_buffers=True)
seek = PassThrough("seek", flush_buffers=True)
truncate = PassThrough("truncate", flush_buffers=True)
flush = PassThrough("flush", flush_buffers=True)
close1 = PassThrough("close1", flush_buffers=True)
try_to_find_file_descriptor = PassThrough("try_to_find_file_descriptor",
flush_buffers=False)
def flushable(self):
return True
class LineBufferingOutputStream(BufferingOutputStream):
"""Line buffering output stream.
This is typically the top of the stack.
"""
def write(self, data):
self.error = False
p = data.rfind('\n') + 1
assert p >= 0
if self.buflen + len(data) < self.bufsize:
if p == 0:
self.buf.append(data)
self.buflen += len(data)
else:
if self.buflen:
self.buf.append(data[:p])
self.do_write(''.join(self.buf))
else:
self.do_write(data[:p])
self.buf = [data[p:]]
self.buflen = len(self.buf[0])
else:
if self.buflen + p < self.bufsize:
p = self.bufsize - self.buflen
if self.buflen:
self.do_write(''.join(self.buf))
assert p >= 0
self.do_write(data[:p])
self.buf = [data[p:]]
self.buflen = len(self.buf[0])
# ____________________________________________________________
class CRLFFilter(Stream):
"""Filtering stream for universal newlines.
TextInputFilter is more general, but this is faster when you don't
need tell/seek.
"""
def __init__(self, base):
self.base = base
self.do_read = base.read
self.atcr = False
def read(self, n):
data = self.do_read(n)
if self.atcr:
if data.startswith("\n"):
data = data[1:] # Very rare case: in the middle of "\r\n"
self.atcr = False
if "\r" in data:
self.atcr = data.endswith("\r") # Test this before removing \r
data = replace_crlf_with_lf(data)
return data
flush = PassThrough("flush", flush_buffers=False)
flushable= PassThrough("flushable", flush_buffers=False)
close1 = PassThrough("close1", flush_buffers=False)
try_to_find_file_descriptor = PassThrough("try_to_find_file_descriptor",
flush_buffers=False)
class TextCRLFFilter(Stream):
"""Filtering stream for universal newlines.
TextInputFilter is more general, but this is faster when you don't
need tell/seek.
"""
def __init__(self, base):
self.base = base
self.do_read = base.read
self.do_write = base.write
self.do_flush = base.flush_buffers
self.readahead_count = 0 # either 0 or 1
def read(self, n=-1):
"""If n >= 1, this should read between 1 and n bytes."""
if n <= 0:
if n < 0:
return self.readall()
else:
return ""
data = self.do_read(n - self.readahead_count)
if self.readahead_count > 0:
data = self.readahead_char + data
self.readahead_count = 0
if data.endswith("\r"):
c = self.do_read(1)
if len(c) >= 1:
assert len(c) == 1
if c[0] == '\n':
data = data + '\n'
else:
self.readahead_char = c[0]
self.readahead_count = 1
result = []
offset = 0
while True:
newoffset = data.find('\r\n', offset)
if newoffset < 0:
result.append(data[offset:])
break
result.append(data[offset:newoffset])
offset = newoffset + 2
return '\n'.join(result)
def readline(self):
line = self.base.readline()
limit = len(line) - 2
if limit >= 0 and line[limit] == '\r' and line[limit + 1] == '\n':
line = line[:limit] + '\n'
return line
def tell(self):
pos = self.base.tell()
return pos - self.readahead_count
def seek(self, offset, whence):
if whence == 1:
offset -= self.readahead_count # correct for already-read-ahead character
self.base.seek(offset, whence)
self.readahead_count = 0
def flush_buffers(self):
if self.readahead_count > 0:
try:
self.base.seek(-self.readahead_count, 1)
except (MyNotImplementedError, OSError):
return
self.readahead_count = 0
self.do_flush()
def write(self, data):
data = replace_char_with_str(data, '\n', '\r\n')
self.flush_buffers()
self.do_write(data)
truncate = PassThrough("truncate", flush_buffers=True)
flush = PassThrough("flush", flush_buffers=False)
flushable= PassThrough("flushable", flush_buffers=False)
close1 = PassThrough("close1", flush_buffers=False)
try_to_find_file_descriptor = PassThrough("try_to_find_file_descriptor",
flush_buffers=False)
class TextInputFilter(Stream):
"""Filtering input stream for universal newline translation."""
def __init__(self, base):
self.base = base # must implement read, may implement tell, seek
self.do_read = base.read
self.atcr = False # Set when last char read was \r
self.buf = "" # Optional one-character read-ahead buffer
self.CR = False
self.NL = False
self.CRLF = False
def getnewlines(self):
return self.CR * 1 + self.NL * 2 + self.CRLF * 4
def read(self, n):
"""Read up to n bytes."""
if self.buf:
assert not self.atcr